Wesley Shrum thinks friendship works better

A decade ago, I began to wonder what would happen if scientists in developing areas could communicate freely and easily with those in Europe.  Could the Internet truly globalize science?

The question was important, because of two striking findings from a 1994 analysis of scientific communication in Ghana, Kenya, and Kerala (India). (See 1.) The first was that the more scientists communicated outside their geographical area, the less they communicated inside.  The second was that scientists from developing countries who were educated in developed countries had no more contact with professionals in developed areas after their return home than those who were educated locally.  One interpretation was that communication difficulties prevented the maintenance of ties, and that their costs required the tradeoff between colleagues inside and outside their area.

At the recent, the dominant view was that collaboration through information and communication technologies will dramatically improve science for developing areas.  While this is conventional wisdom at the present time, our evidence suggests it is time to reexamine this view.

Benefits and costs
The benefits of collaboration seem clear.  Institutions collaborate when they do not themselves possess the funds, equipment, or personnel necessary to achieve some objective.  They collaborate when multilateral or national funding centres establish a programme and when their leadership recognizes an opportunity, sees a trend, and commits resources.  Researchers collaborate to solve technical and scientific problems, but they do so in the context of careers and organizations that are very different in Africa, Latin America, and Asia.

When people decide to collaborate, they incur costs as well as benefits.  During the past decade the benefits have been emphasized while the costs have been largely ignored.  The promise of the Internet is that the reduction of communication costs makes coordinating projects easy, changing the overall balance of costs and benefits.  Post-Internet collaboration should be easy and productive because costs are low and benefits are great.

Surprising findings
Unfortunately, that is not the case in the developing world.  For the past decade we have examined the changes wrought by the Internet in the scientific communities of developing areas. (See 2)  There is little evidence that Internet collaboration generally increases productivity. Sometimes it seems to reduce output. 

Our Indian scientists, who are relatively productive, collaborate very little.  Kenyan scientists, who collaborate a great deal, are relatively unproductive.  As a Ugandan put it during the recent World Summit on the Information Society, ‘Africans love to start projects.  They just don’t like finishing them.’  But a Filipino administrator put it more accurately: scientists have got to move on to the next project or consultancy as soon as the final report is written.  The need to publish the results in the open literature is secondary, and often they are ill-prepared to do it.

There are two major aspects to the communication problem.  One is that many research institutes and universities in developing areas are simply not connected to the Internet, when ‘connectivity’ is understood as an unshared computer in one’s office with a broadband (always on) connection to the Internet.  The second is that scientists, policy makers, and programme managers assume that their colleagues in the developing world are connected to the Internet and use it in the same way (‘sure, there are some problems – but I get emails from Africa all the time’).  The evidence indicates that costs are indeed lower than before, but access to the Internet is limited, shared, and often domestic.

Limited change
The Internet has changed the conditions under which data are stored and manipulated. It has altered the speed of communication, but it has not changed the local processes that shape the way researchers are hired and paid, the way they teach, their extended family obligations, their administrative responsibilities, consulting opportunities, and dependency on donors.

Seeking to understand international scientific collaboration, we have had to form one – and that has changed us in the process.  Perhaps the most important aspect is friendship, from which I have learned that lower Internet connectivity, and different ways of using the Internet, is not actually a ‘problem’ except as I defined it as such.  It is best viewed as a condition.  Donors – like me – have become a significant part of the long term problem by creating conditions of dependency.  Social and scientific relationships across national boundaries are the only antidote.  And they are not best created through the Internet.

References
1. "Are Scientists in Developing Countries Isolated?"  2000. Wesley Shrum and Patricia Campion. Science, Technology, and Society 5(1): 1-34.
2. The original sites were Ghana, Kenya and Kerala (India).  More recently we have studied South Africa, Chile and the Philippines.

Wesley Shrum is Professor of Sociology at Louisiana State University